R Leaflet : detect on which polygon is the map bounds center - r

Question

I would like to automatically detect which polygon is at the center of the map. And it should update dynamically when the user is moving through the map.
For the moment I could not find a way to reverse find on which polygon are some coordinates.
I think I could simulate a input$map_shape_click with shinyjs or javascript and so get input$map_shape_click$id, but before I go to this solution, I would like to make sure there is no other way.
Here is a minimal example
library(leaflet)
library(shiny)
# data source : https://biogeo.ucdavis.edu/data/gadm3.6/Rsp/gadm36_FRA_2_sp.rds
cities <- readRDS(file = "../gadm36_FRA_2_sf.rds")
ui <- fluidPage(leafletOutput("map"))
server <- function(input, output, session) {
rv <- reactiveValues()
output$map <- renderLeaflet({
leaflet() %>%
addProviderTiles(provider = providers$CartoDB.Positron) %>%
setView(lng = 1, lat = 45, zoom = 8) %>%
addPolygons(data = cities,layerId = ~NAME_2,label = ~NAME_2)
})
observeEvent(input$map_bounds,{
rv$center <- c(mean(input$map_bounds$north, input$map_bounds$south), mean(input$map_bounds$east, input$map_bounds$west))
# how can I detect on which polygon the center is ?
})
}
shinyApp(ui = ui, server = server)

Answer 1

library(leaflet)
library(shiny)
library(sf)
cities <- readRDS(file = "gadm36_FRA_2_sp.rds") %>%
st_as_sf()
ui <- fluidPage(leafletOutput("map"))
server <- function(input, output, session) {
rv <- reactiveValues()
output$map <- renderLeaflet({
leaflet() %>%
addProviderTiles(provider = providers$CartoDB.Positron) %>%
setView(lng = 1, lat = 45, zoom = 8) %>%
addPolygons(data = cities, layerId = ~NAME_2, label = ~NAME_2)
})
observeEvent(input$map_bounds, {
rv$center <- c(mean(input$map_bounds$north, input$map_bounds$south), mean(input$map_bounds$east,
input$map_bounds$west))
pnt <- st_point(c(rv$center[2], rv$center[1]))
rslt <- cities[which(st_intersects(pnt, cities, sparse = FALSE)),]$NAME_1
print(rslt)
})
}
shinyApp(ui = ui, server = server)

Answer 2

So I found a way to do it with the function sf::st_intersects
observeEvent(input$map_bounds,{
rv$center <- data.frame(x = mean(c(input$map_bounds$north, input$map_bounds$south)),
y = mean(c(input$map_bounds$east, input$map_bounds$west)))
res <- sf::st_as_sf(rv$center, coords=c("y","x"), crs=st_crs(cities$geometry))
intersection <- as.integer(st_intersects(res, cities$geometry))
print(if_else(is.na(intersection), '', cities$NAME_2[intersection]))
})